The provision of online traffic information and applications in vehicular environment can be offered via fixed ground Roadside Units (RSUs). Nevertheless, employing flying RSUs that are carried by Unmanned Aerial Vehicles (UAVs) would bring the capabilities of Mobile Edge Computing (MEC) to the vehicular environment, where they can be deployed dynamically in accordance to traffic safety events and congestion conditions. In this paper, we propose a novel method for the deployment of flying RSUs based on swarm intelligence. Our proposed method relies on two fundamental properties of the dynamic deployment of edges in vehicular environments, which are self-organization and the division of workload among RSUs. These two properties are necessary and sufficient to obtain a swarm intelligent behavior. The term 'swarm' can refer to any restrained collection of interacting agents or individuals, which are - in this case - the travelling vehicles and RSUs.